Documents
Document → Excel
Convert business documents into structured Excel data.
Document → Excel reads the documents your business works with, picks out the information you need and delivers it as structured rows in Excel or a database. Uncertain values are flagged for a person to check.
What it does
Many business documents contain the information you need, but not in a form you can sort, filter or reuse. Document → Excel reads PDFs, Word files, scanned pages and images, and extracts the fields you define: names, dates, reference numbers, amounts, line items or anything else your process depends on.
Each value is stored as a structured field together with a confidence level. Values that need attention are shown next to the original document, so a person can confirm or correct them before the data is exported.
Typical input
- PDF documents
- Word documents
- Scanned documents and images
- Spreadsheet-based source files, where relevant
What you receive
- Structured fields defined for your document types
- An Excel file in the layout you work with
- Records stored in a database for later use
- A clear view of values that were reviewed or corrected
Works on its own
Document → Excel is sold and installed on its own. If you later use other products, they can be connected as an optional customization.
The problem it solves
In many companies, information from documents is still copied into spreadsheets by hand. The work is repetitive, it takes time away from more useful tasks, and small typing errors are easy to miss.
- Staff retype the same fields from PDFs and Word files every day
- Scanned documents cannot be searched, sorted or filtered
- Typing errors find their way into reports and other systems
- Useful information stays spread across individual files
How it works
- 1
Upload documents
Documents are added through a simple web interface, individually or in batches.
- 2
Read the content
Text is read directly from digital files. Scanned pages and images are processed with OCR.
- 3
Extract the fields
AI-assisted extraction fills in the fields defined for each document type.
- 4
Review where needed
Each value has a confidence level. Uncertain or missing values are flagged for a person to check.
- 5
Export the data
Approved data is exported to Excel and stored in a database.
Example workflow
A company keeps a register of supplier certificates it receives as PDF and Word files, and occasionally as scans.
- 01An employee uploads this week's certificates in one batch
- 02Supplier name, certificate type, reference number, issue date and expiry date are extracted from each document
- 03An expiry date on a poorly scanned page is flagged as uncertain
- 04The employee checks it against the original and confirms the value
- 05The new records are added to the company's Excel register
Key capabilities
PDF and Word documents
Reads digital PDFs and Word files directly.
Scans and images
Uses OCR to read scanned pages and photographed documents.
Configurable fields
Extracts the fields defined for each of your document types.
Confidence handling
Shows how reliable each extracted value is.
Human review
Flagged values are checked side by side with the source document.
Excel and database output
Exports to Excel and keeps structured records in a database.
Expected business impact
- Reduce repetitive copy-and-paste work from documents
- Turn unstructured documents into data you can sort, filter and reuse
- Keep information in a consistent structure across all documents
- Focus manual checking on the values that actually need it
Before
- Open each document and look for the relevant values
- Type the values into a spreadsheet by hand
- Check the whole spreadsheet for mistakes
After
- Upload documents in a batch
- Check only the values flagged for review
- Export a structured Excel file
The actual effect depends on your volumes and current process. We discuss realistic expectations during the requirements phase.
Customization
Core automation, adapted to your workflow.
The extraction and review process is the same for every customer. What changes is how it is set up for your documents and your output.
- Document types and the fields extracted from each
- Review rules and when a value is flagged
- Excel layout and column structure
- Database structure for stored records
- Connections to other systems, where agreed in the scope
Ownership & deployment
This product is installed into infrastructure that your organization owns and pays for. After delivery, you remove our access and keep full control.
- Your repository, database and deployment
- Your data, API accounts and external service accounts
- Developer access only during the project, removed by you after delivery
- Keeps running without our accounts or servers
| GitHub | Private repository with the system's code |
|---|---|
| Supabase | Database and document storage |
| Vercel | Hosting for the upload and review interface |
| Anthropic API | AI-assisted reading and extraction |
Pricing
Project price
€1,800 – €2,500
A one-time project price for customization, installation, testing and delivery. The final price is agreed in writing after the requirements discussion.
Founding Client PricingIntroductory pricing for our first 3 client projects.
Prices exclude VAT where applicable.
After successful delivery, we may ask for permission to use anonymized results or a case study. This is always your choice.
Included in the project
- Setup for the agreed document types and fields
- Installation in your infrastructure
- Testing with your own documents
- Handover and documentation
What affects the final price
- Number of document types
- Number and complexity of fields
- Share of scanned or low-quality documents
- Output requirements beyond standard Excel export
Running costs of services in your own accounts (hosting, database, AI API usage) are billed to you directly by those providers.
How to order
Every product is ordered the same way:
- 1Contact
- 2Requirements
- 3Scope & quote
- 4Customization
- 5Implementation
- 6Testing
- 7Delivery
Talk to us about Document → Excel
Send a short description of your current workflow. We will reply with questions for scoping and the next steps.